INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

The plastid DNA contains 11 ndh genes (Maier et al., 1995) encoding polypeptides (NDH) that are components of the plastid Ndh complex, analogous to the NADH dehydrogenase or complex I (EC 1.6.5.3) of mitochondrial respiratory chain (Sazanov et al., 1998; Casano et al., 2000). The increases of NDH polypeptides and NADH dehydrogenase activity of the Ndh complex (NADH-DH) under photooxidative stress (Martín et al., 1996; Casano et al., 1999, 2000) suggest that the Ndh complex is involved in the protection against such stress. In fact, ndh-less mutants show increased sensitivity to photooxidative stress (Endo et al., 1999; Horvath et al., 2000). The purified Ndh complex catalyzes the transfer of electrons from NADH to plastoquinone and, in vivo a thylakoid plastoquinol peroxidase probably oxidizes the reduced plastoquinone with H₂O₂ (Casano et al., 2000). Ndh complex (providing electrons) plus plastoquinol peroxidase with Mehler reaction and superoxide dismutase (draining electrons) might poise the redox level of the electron carriers. This mechanism (chlororespiration) would most likely ensure the photosynthetic electron transport under a variety of environmental conditions that include rapid changes of light intensity associated with sunflecks and leaf movements. In addition, the chlororespiration may act as system scavenging reactive oxygen species generated under continuous photooxidative stress or by the successions of sunflecks and light gaps (Casano et al., 2000).

The increase in the levels of NDH polypeptides and Ndh complex activity (Martín et al., 1996; Casano et al., 1999, 2000) is the first described case of plastid DNA-encoded proteins that are stimulated by photooxidative stress. Thus, even though assuming an initial control of the plastid-targeted actions at the level of nucleus-cytoplasmic system, it is of interest to investigate whether or not H₂O₂ generated in the chloroplast could mediate the increase in the level of plastid-encoded proteins.

The photooxidative stress response shares strong similarities with the response of plants to pathogens (Levine, 1999), where a still poorly understood signal transduction pathway includes H₂O₂ and salicylic acid as components. Increasing evidence (for review, see Levine, 1999) suggests that the high concentrations of superoxide anion radical and H₂O₂ in the infection focus are high enough to kill not only the pathogen, but also the targeted plant cells (hypersensitive response). In these cells the concentration of H₂O₂ would trigger a programmed cell death. Meanwhile, the concentration of H₂O₂ in neighboring cells, at appropriate distance, would reach a lower level, which induces a succession of protecting genes that encode the pathogenesis-related proteins. The response of leaves to increasing photooxidative stress is very similar (Casano et al., 1999). At low photooxidative stress the battery of enzymes scavenging or preventing reactive oxygen species accumulation is induced. However, at higher photooxidative stress even these protective enzymes are more rapidly destroyed than induced (Casano et al., 1999). Moreover, H₂O₂ seems to be involved in the oxidative stress-mediated induction of nuclear-encoded defensive enzymes such as cytosolic ascorbate peroxidase (Karpinski et al., 1999; Morita et al., 1999), glutathione S-transferase, and catalase (Polidoros and Scandalios, 1999). Significant age-dependent differences in the response of protective enzymes to increasing photooxidative stress (Casano et al., 1999) also suggest a close relation among the mechanisms involved in photooxidative stress response, pathogen defense response, leaf cell senescence, and hypersensitive response.

Bearing in mind the above mentioned similarities we have investigated the possible involvement of H₂O₂ in the increase of plastid NDH polypeptides and NADH-DH under photooxidative stress. We have also investigated the possibility of whether these increases are related to the increases of the levels of the corresponding chloroplastic mRNAs.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Expression of Plastid ndh Genes in Response to Photooxidative Stress

Most of ndh genes are transcribed as policistrons in a fashion similar to a number of chloroplast-encoded genes (del Campo et al., 2000). However, ndhB and ndhF are transcribed monocistronically (Martínez et al., 1997) and their transcripts were detected at their predicted size, 1,650 and 2,400 b (data not shown), respectively, in freshly detached leaves of 7- and 14-d-old barley (Hordeum vulgare cv Hassan) plants (Fig. 1). The steady-state level of ndhB transcripts decreased or did not change after darkness or under growing light (GL) conditions, whereas ndhF transcripts slightly increased under GL in both types of leaves. In contrast, a marked increase of the steady-state level of both transcripts was observed in response to 20-h incubation under relatively excess light (photooxidative light [PhL]). However, the initial as well as the photooxidative-induced levels of both transcripts were significantly higher in 14-d-old leaves than in expanding 7-d-old leaves. This pattern of ndh transcripts correlates with parallel changes in the amount of NDH-F protein and the activity of the Ndh complex (Casano et al., 1999).

View larger version (60K): [in this window] [in a new window]   Figure 1.   Changes in the steady-state transcript levels of the plastid ndhB and ndhF genes in response to different light treatments. Primary leaves from young (7-d-old) and mature-senescent (14-d-old) plants were incubated for 20 h at 23°C in darkness (D), GL (100 µmol photon m2 s1), or PhL (300 µmol photon m2 s1). Total RNA was isolated from barley leaves, separated by agarose electrophoresis (12.5 µg each), blotted on a nylon membrane, and hybridized with ndhB or ndhF probes or stained with methylene blue (for rRNAs) as described in "Materials and Methods."

Effects of Hydrogen Peroxide on the Expression of ndh Genes

Generation of H₂O₂ during photooxidative stress has been proposed as a part of the signaling cascade leading to induction of nuclear-encoded protecting enzymes (Morita et al., 1999; Polidoros and Scandalios, 1999). It was interesting to investigate whether or not H₂O₂ is involved in the induction of Ndh complex, which participates in the protection of chloroplasts against photooxidative stress (Casano et al., 1999, 2000). As a consequence, a study was carried out on 14-d-old leaves due to their increased response to photooxidative treatments as stated above.

Changes in the Activity of the Ndh Complex and in the Level of NDH-F Protein

View larger version (31K): [in this window] [in a new window]   Figure 2.   Effect of hydrogen peroxide on the Ndh complex activity and NDH-F protein. Typical zymogram of plastid NADH-DH and western blot with antibody against NDH-F of the Ndh complex from 14-d-old leaves, including activities of freshly detached leaves (T0) and of leaves incubated at 23°C for 20 h with indicated concentrations of H2O2 in GL or darkness. For zymograms, 50 µg of protein and for protein blots, 25 µg of protein of leaf crude extracts was loaded per lane. Detailed procedures are described in "Materials and Methods."

View larger version (29K): [in this window] [in a new window]   Figure 3.   Effects of DMTU, hydrogen peroxide, and photooxidative stress on Ndh activity. NADH-DH of Ndh complex was deduced from zymograms (not shown) of crude extracts (50 µg of protein per lane) from 14-d-old leaves pre-incubated with 0 and 5 mM DMTU for 4 h at 23°C under 100 µmol photon m2 s1. Thereafter, leaves were transferred to GL, GL and 5 mM H2O2 in the incubation medium (GL + H2O2), PhL and PhL and 50 nM PQ in the incubation medium (PhL + PQ) for 20 h at 23°C. Activities were expressed as percentages of the values in freshly detached leaves (7 nmol NADH oxidized min1 mg1 protein). Values are means of four different experiments.

View larger version (29K): [in this window] [in a new window]   Figure 4.   Time-dependent induction of plastid Ndh complex activity and NDH-F protein in photooxidative stressed leaves. A, Typical zymogram of NADH-DH from 14-d-old leaves, including activities of leaves immediately after pre-incubation with 0 and 5 mM DMTU for 4 h at 23°C under 100 µmol photon m2 s1 (Time, 0), and after transfer to GL, PhL, GL and 5 mM H2O2 in the incubation medium (GL + H2O2), and PhL and 50 nM PQ in the incubation medium (PhL + PQ) up to 30 h at 23°C. B and C, Specific NADH dehydrogenase activities of Ndh complex were deduced from experiments as those of A. D and E, Level of NDH-F protein of Ndh complex relative to total soluble protein in leaves treated as in A and deduced from western blots with antibody against NDH-F. For zymograms, 50 µg of protein and for protein blots, 25 µg of protein of leaf crude extracts was loaded per lane. In B through E, activities and protein were expressed as percentages of the values in leaves immediately after pre-incubation (T0). Detailed procedures are described in "Materials and Methods." Values are means of at least three different experiments.

Changes in the Level of ndhB and ndhF Transcripts

View larger version (41K): [in this window] [in a new window]   Figure 5.   Effects of DMTU, hydrogen peroxide, and photooxidative stress on the steady-state levels of plastid ndhB transcripts. A, Northern blot of total RNA from 14-d-old leaves, including RNA of freshly detached leaves (T0) and RNA of leaves pre-incubated with 0 and 5 mM DMTU for 4 h at 23°C under 100 µmol photon m2 s1, and then transferred to GL and 5 mM H2O2 in the incubation medium (GL + H2O2), GL and PhL, and PhL and 50 nM PQ in the incubation medium (PhL + PQ) at 23°C for the indicated times. Total RNA was isolated from barley leaves, separated by agarose electrophoresis (12.5 µg each), blotted on a nylon membrane, and hybridized with ndhB probe as described in "Materials and Methods." B and C, Relative level of ndhB transcripts was deduced from experiments as those of A and normalized to the respective ribosomal RNA. Values were expressed as percentages of those in freshly detached leaves and represent the mean of three experiments.

View larger version (48K): [in this window] [in a new window]   Figure 6.   Effects of DMTU, hydrogen peroxide, and photooxidative stress on the steady-state levels of plastid ndhF transcripts. A, Northern blot of total RNA from 14-d-old leaves, including RNA of freshly detached leaves (T0) and RNA of leaves pre-incubated with 0 and 5 mM DMTU for 4 h at 23°C under 100 µmol photon m2 s1, and then transferred to GL and 5 mM H2O2 in the incubation medium (GL + H2O2), GL, PhL, and PhL and 50 nM PQ in the incubation medium (PhL + PQ) at 23°C for the indicated times. Total RNA was isolated from barley leaves, separated by agarose electrophoresis (12.5 µg each), blotted on a nylon membrane, and hybridized with ndhF probe as described in "Materials and Methods." B and C, Relative level of ndhF transcripts was deduced from experiments as those of A and normalized to the respective ribosomal RNA. Values were expressed as percentages of those in freshly detached leaves and represent the mean of three experiments.

Effects of Hydrogen Peroxide on Thylakoid Peroxidase

The thylakoid peroxidase can scavenge H₂O₂ by oxidizing the plastoquinol, which has been reduced by the action of Ndh complex (Casano et al., 2000). Like Ndh complex, plastoquinol peroxidase activity increases under photooxidative stress, especially in mature-senescent leaves (Casano et al., 1999). Therefore, it was interesting to study whether or not H₂O₂ is also involved in the induction of the thylakoid peroxidase activity. Time-dependent changes of peroxidase activity were studied in leaves pre-incubated with or without DMTU and then incubated under GL, PhL, and GL + H₂O₂ up to 30 h at 23°C. Typical peroxidase zymograms are shown in Figure 7A. Incubation under GL did not change peroxidase activity significantly in water- and DMTU-pretreated leaves (Fig. 7, B and C, respectively). However, in water-pretreated leaves PhL caused a sharp increase of the enzymatic activity after a 4-h incubation, reaching a 3-fold level of the initial value and it then stabilized (Fig. 7B). The early inductive effect was also triggered by incubation with H₂O₂ (GL + H₂O₂), but after a 16-h incubation peroxidase activity did not differ from that of control. When leaves were pretreated with DMTU, no inductive effect of PhL or GL + H₂O₂ on peroxidase activity was observed (Fig. 7C). In general, the obtained results are in agreement with previous observations (Casano et al., 1999, 2000) and suggest that photooxidative induction of thylakoid peroxidase precedes in time that of Ndh complex.

View larger version (35K): [in this window] [in a new window]   Figure 7.   Time-dependent induction of thylakoid peroxidase activity in photooxidative stressed leaves. A, Typical zymogram of thylakoid peroxidase activity from 14-d-old leaves, including activities of leaves immediately after pre-incubation with 0 and 5 mM DMTU for 4 h at 23°C under 100 µmol photon m2 s1 (T0), and after transfer to GL, PhL, and GL and 5 mM H2O2 in the incubation medium (GL + H2O2) up to 30 h at 23°C. B and C, Specific thylakoid peroxidase activities were deduced from experiments as those of A. Fifty micrograms of protein of leaf crude extracts was loaded per lane. In B and C, activities were expressed as percentages of the values in leaves immediately after pre-incubation (T0; 70 nmol HQ oxidized min1 mg1 protein). Detailed procedures are described in "Materials and Methods." Values are means of at least three different experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

The ndhB and ndhF genes are transcribed from monocistronic units encoded in the inverted repeated and the small single-copy region of plastid DNA, respectively (Freyer et al., 1995; Maier et al., 1995; Martínez et al., 1997). On the other hand, six (H, A, I, G, E, and D) and three (C, K,, and J) ndh genes are respectively grouped within two polycistronic transcriptional units that produce multiple transcripts, probably by a complex processing of primary transcripts (Maier et al., 1995; del Campo et al., 2000). The first unit also includes the psaC gene encoding a polypeptide of the photosystem I between ndhE and ndhD. There are uncertainties about which transcripts of each of the two polycistronic units are translated. Preliminary assays in our laboratory indicate that photooxidative and hormone treatments affect the steady-state levels of the different transcripts of polycistronic units in a complex way, suggesting that post-transcriptional processing may be involved in the control of the NDH polypeptides synthesis encoded in polycistronic units. In contrast, the effects of leaf treatments on the levels of the single transcripts of ndhB and ndhF genes can be more easily investigated. Figures 1, 5, and 6 show that ndhB and ndhF transcripts increased after treatments producing oxidative stress, especially in mature-senescent leaves. This correlates well with increases of NDH polypeptides and NADH-DH of Ndh complex under photooxidative conditions (Martín et al., 1996; Catalá et al., 1997; Casano et al., 1999, 2000). All together, the results indicate that the induction of plastid Ndh complex is mediated, at least in part, by increases of mRNA levels.

Even though photooxidative stress is initiated within the chloroplasts, the unscavenged excess of H₂O₂ can rapidly diffuse out of the plastid (directly or indirectly), generating a situation of high risk of oxidative damage for the entire cell. To orchestrate an effective cell protection, a number of nuclear- and chloroplast-encoded genes must be induced coordinately. It is possible that some common signaling intermediate is involved in the antioxidant response at chloroplastic and nuclear levels. Environmental stresses such as suboptimal or extreme temperatures (Prassad et al., 1994; Fadzillah et al., 1996; Dat et al., 1998) and excess light (Karpinski et al., 1997) are known to increase the steady-state levels of the H₂O₂. The photooxidative protecting Cat and Gst1 nuclear-encoded genes (Polidoros and Scandalios, 1999) are induced by H₂O₂ in maize in a dose- and time-dependent fashion. Kovtun et al. (2000) recently demonstrated that a specific H₂O₂-responsive mitogen-activated protein kinase cascade mediates the H₂O₂ induction of Gst expression. A direct signaling action of H₂O₂ has also been described (Karpinski et al., 1999; Morita et al., 1999) for the nuclear-encoded cytosolic ascorbate peroxidase. In a similar manner, the incubation of barley leaf segments with H₂O₂ increased NADH-DH (Figs. 2 and 3), NDH-F polypeptide (Fig. 2), and ndhB and ndhF transcripts (Figs. 5 and 6). H₂O₂ applied under GL mimicked the effects of PhL and PhL + PQ, and all these effects were suppressed when leaf segments were pre-incubated with the H₂O₂ scavenger DMTU. These results strongly suggest a direct signaling of H₂O₂ in the induction of protective response against photooxidative stress in chloroplasts. Thus, H₂O₂, which is diffusible through membranes and, at least under photooxidative conditions, is mainly generated in the chloroplasts, induces the expression of specific nuclear and plastid genes. One may wonder whether its action on the expression of plastid ndh genes would depend on previous action(s) at the nucleus-cytoplasmic compartments.

Although increases of NADH-DH and NDH-F polypeptide after 10 to 30 h of H₂O₂ or photooxidative treatments (PhL and PhL + PQ; Fig. 4) may be mainly due to increases of transcript levels, the 100% increase of NDH-F and the 50% to 80% increase of NADH-DH after 4 h of treatment were not accounted for by increases of ndhB and ndhF transcripts (Figs. 5 and 6), except for PhL + PQ treatment. This suggests that in addition to the effect mediated by increasing mRNAs levels, photooxidative treatment and H₂O₂ (as its transduction signal) promote the translation of pre-existing transcripts of ndh genes. It is significant that an increase in NDH-F polypeptide and NADH-DH under oxidative treatment (Fig. 4, B and D) showed a two-phase behavior, the first one probably due to an effect on the translatability and the second one to an effect on mRNA level.

On the other hand, the relatively small, but significant increase of transcripts after 4 h of incubation with PhL + PQ (Figs. 5B and 6B) suggests that the intrachloroplastic H₂O₂ produced under such conditions could modulate transcript levels without the involvement of newly synthesized nucleus-cytoplasmic intermediates. However, the rapid increase of thylakoid peroxidase (Fig. 7, A and B), which is presumably encoded in the nucleus, also suggests a rapid action of H₂O₂ in the nucleus-cytoplasmic compartments, as reported for other H₂O₂-scavenging enzymes (Karpinski et al., 1999; Morita et al., 1999; Polidoros and Scandalios, 1999). An alternative explanation is that the thylakoid peroxidase could be activated directly by H₂O₂.

The enhanced expression of ndh genes reported in this paper is first described for plastid DNA genes under photooxidative stress. Plastid DNA from several species has been completely sequenced and although the control plastid gene expression includes different post-transcriptional steps (del Campo et al., 2000; Mayfield et al., 1995), its low size makes it a good candidate for investigating the induction of gene expression under photooxidative stress. The involvement of H₂O₂ as a key component of the signal transduction pathway in the responses of nucleus and chloroplast DNAs suggests that some similar steps or mechanisms coordinate the responses of the two genomes against photooxidative stress.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Plant Material

Barley (Hordeum vulgare cv Hassan) was grown on vermiculite in a controlled growth chamber at 23°C under a 16-h photoperiod of 100 µmol photon m² s¹ white light as described by Casano et al. (1999). In the present work we have used primary leaves of 7- and 14-d-old plants as young expanding and aged-senescent leaves, respectively. Subapical leaf segments (3 cm in length) were cut 4 to 5 h after the beginning of photoperiod and were incubated at 23°C up to 30 h with different concentrations of H₂O₂ or 50 nM PQ in darkness, GL (100 µmol photon m² s¹), or PhL (300 µmol photon m² s¹). In some experiments the leaf segments were treated with 0 or 5 mM DMTU for 4 h in GL at 23°C prior to treatment of leaves as described above.

RNA Isolation and Northern-Blot Analysis

Total RNA of leaf segments (2 g) was extracted by phenol-SDS treatment and selective precipitation with LiCl as described (Jones et al., 1985; Eker and Davies, 1987). It was typical that RNA yields were around 0.25 mg/g leaves. RNA samples (12.5 µg) were denatured in formaldehyde and run on 1.2% (w/v) agarose-18% (v/v) formaldehyde gels (Sambrook et al., 1989). After electrophoresis, denatured RNA was immobilized on nylon membranes (Zeta-Probe, Bio-Rad, Hercules, CA). Ribosomal RNAs and M_r markers (Boehringer Mannheim, Mannheim, Germany) were stained with methylene blue (Sambrook et al., 1989) and then scanned using an UVP Easy digital image analyzer (Ultra-Violet Limited, San Gabriel, CA), with automatic background correction. Thereafter, membranes were hybridized to digoxigenin-labeled PCR probes of ndhB and ndhF. Washings were performed under high stringency conditions. Transcript bands were scanned and mRNA levels were expressed on a total rRNA basis as described (Casano et al., 1994).

Homologous digoxigenin-labeled PCR probes of ndhB and ndhF genes were prepared (Lo et al., 1990) from barley plastid DNA as template. Barley plastid DNA was isolated as described by Heinhorst et al. (1988). The probes were enriched in the strand complementary to transcripts by using a 50:1 ratio of 3' terminal:5' terminal primer for each corresponding mRNA. In this order the used primers were: ATCGATTCAACCTCTGAT and AGCCTCATTAGACCGTAG spanning a 396-bp of ndhB probe in barley (Freyer et al., 1995) and CCCACAGTAACTACCT and GCGTTTTATATGTTTCGG spanning a 735-bp ndhF probe near the 3' end in rice (Hiratsuka et al., 1989) and probably in barley.

Preparation of Leaf Crude Extracts

For zymographic and western-blot assays, activities and proteins were assayed in whole-leaf extracts obtained as follows: 10 leaf segments were homogenized with a mortar and pestle in 2 mL of 50 mM potassium phosphate, pH 7.0, 1 mM L-ascorbic acid, 1 mM EDTA, and 5% (w/v) polyvinylpirrolidone, and were centrifuged at 500g for 10 min. Triton X-100 was added to supernatant to make a final 2% (w/v) solution and gently stirred for 30 min. The suspension was centrifuged at 20,000g for 30 min. Supernatants contained 0.7 to 1.3 mg protein mL¹. The entire procedure was carried out at 4°C.

Gel Electrophoresis, Zymograms, and Immunoassays

Native PAGE was carried out at 5°C (usually with 100 µg of protein samples) in a linear gradient gel of 3% to 10% (w/v) polyacrylamide (2.5% [w/v] bis-acrylamide) in the same way as SDS-PAGE with the exception that gels contained 0.1% (w/v) Triton X-100 instead of SDS (Casano et al., 1999). For zymograms, NADH-DH of Ndh complex was detected by incubation of gel slices for 20 to 30 min at 30°C in darkness in 50 mM potassium phosphate (pH 8.0), 1 mM Na₂-EDTA, 0.2 mM NADH, and 0.5 mg mL¹ nitroblue tetrazolium. In controls without NADH, no stain developed. Staining for peroxidase was performed by following standard methods with 4-methoxy--naphthol, as described by Casano et al. (1999). For immunoblot analyses, after SDS-PAGE, proteins were transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA). The immunocomplex with antibodies prepared against the NDH-F polypeptide encoded by the ndhF gene (Catalá et al., 1997) was detected with the alkaline phosphatase western-blotting analysis system (Boehringer Mannheim).

Bands from zymograms and immunoblots were scanned with a UVP Easy Digital Image analyzer to comparatively quantify activity and protein values, which were expressed as percentages of the reference (that of freshly detached primary leaves).

Other Determinations

The NADH:FeCN oxidoreductase activity, specific for Ndh complex, was assayed at 30°C by measuring the reduction of FeCN at 420 nm and the oxidation of NADH at 340 nm as described (Casano et al., 2000). The spectrophotometric assay of thylakoid peroxidase was performed at 30°C by measuring the oxidation rate of hydroquinone in the presence of H₂O₂, as described by Casano et al. (1999). Specific activities were expressed as micromoles of NADH or hydroquinone consumed per minute per milligram of protein.

Protein concentration was quantified by the Bradford method (1976) with a Protein Assay Kit (Bio-Rad) using bovine serum albumin as a standard.

All reported results were reproduced at least three times. When appropriate, standard deviations were indicated by bars in figures.